Gathering winds of change

When I mention that I used to write for MacUser, most people used to find it curious, even a little eccentric, that I should use and develop software for the Mac. It was a bit like admitting that you are a Zen Buddhist (which I am not), or wearing sandals with a suit (which I do, often).

Now almost everyone who I tell about MacUser launches into a soliloquy about how they haven’t looked back since switching to a Mac at home or, for the most enlightened, at work. They still stare at the sandals, of course.

Although Apple’s recent runaway success has been fuelled by i-series products including iPods, iPhones, iTunes, and iPads, it has also become one of the most successful computer manufacturers. Macs have broken out of their design and publishing niche to occupy central positions in very many homes, schools, and businesses. But in contrast to the iPhone, with a new model every year, and the iPad, remodelled every 8 months or so, Macs don’t seem to have changed much for quite a few years.

Classic Macs had characteristic architectural features built around Motorola 68K processors, and were replaced by Power Macs after 10 years. The PowerPC was overthrown by Intel x86 processors with their own supporting chipsets and buses in 2006, 9 years ago. Only two of the current Mac range, the 2013 Mac Pro and the latest MacBook, have undergone significant redesign in the last 7 years, since the release of the MacBook Air.

Apple’s Classic operating system family had a lifespan of 17 years, until Mac OS X 10.0 launched in March 2001, 14 years ago. We are now well over half way through OS X’s likely lifespan, which means that Apple should already be well engaged in developing its successor.

So is Apple pressing ahead to develop OS XI and new hardware ranges, or is it going to leave its computers to wither on the vine?

One way of reading where Apple is heading is to study its technology acquisitions and partners. On the hardware side its interests are strongest in SOC (‘system on a chip’) development, which has so far seen greatest application in iPhones and iPads.

But the SOC approach has merits for more general computer design, in shrinking the volume required by hardware even further. Coupled with solid-state disks and externalisation of mass storage to Thunderbolt 2 peripherals, we could soon be using an ‘iMac Air’ which consists of little more than a display and input devices: the logical companion to the new Mac Pro.

As the CPU manufacturers seem best able to give us more rather than faster cores, the OS XI kernel needs to be even better at distributing and handling multiple processes over those cores, and could change radically from the current ageing descendent of Mach, perhaps towards an L4 derivative, or a third generation microkernel.

Apps too must be more finely granular so that they can spread over the available cores and resources such as GPUs to get greatest benefit. This will not be as radical a change as the switch from Classic to OS X, but sooner or later Apple will need to draw a line under OS X and bring a new OS to market.

It is helpful to Apple that press attention has been largely focussed on its high-profile i-series activities, its new consumer product, the Watch, and speculation about TVs and cars. But if it wants to see continuing returns from its quietly successful and less volatile computer products, Apple will have several hundred engineers pushing forward the specifications for the next generation. Now is the time to watch for clues as to what they are up to.

This has been updated from the original, published in MacUser vol 30 issue 08, 2014.